Social Phobia is a highly prevalent, disabling, and chronic anxiety disorder associated with substantial vocational, social, and academic impairment. It increases the risk of depression, substance abuse and financial dependence. Treatment for social phobia has received less attention than treatment for other anxiety disorders. However, specific psychosocial and psychopharmacological treatments have demonstrated efficacy. Cognitive Behavioral Group Therapy (CBGT), developed by Heimberg, has proven superior to control therapies while the MAOI phenelzine (PZ), first studied by Liebowitz, has proven superior to placebo in several controlled trials. In a novel and highly successful collaboration, Heimberg and Liebowitz together compared PZ, CBGT, pill placebo, and a credible psychosocial control, and found PZ and CBGT superior and of distinct clinical benefit. PZ was faster acting and more effective than CBGT on some measures while CBGT effects seemed more durable following treatment discontinuation. While PZ and CBGT produced positive effects, many patients did not respond enough, and PZ discontinuation led to significant relapse. Therefore, we still need treatments that are more effective and durable than CBGT or PZ as typically administered. We have recently been funded to study acute efficacy of combined PZ/CBGT, which may be more effective than PZ or CBGT alone. programmatic budget cuts, however, have prevented study of long-term efficacy of CBGT, Pz, or the combination. CBGT needs further study to determine whether longer, intensive treatment results in greater gains. PZ needs further study to determine if longer treatment reduces relapse after PZ discontinuation. Long-term study of PZ/CBGT is needed to determine if the combination fuses strengths of each component, i.e., efficacy of PZ and durability of CBGT. If so, combined PZ/CBGT treatment would be of distinct advantage to many patients. We now propose to examine the long-term efficacy and durability of CBGT, PZ, and the combination in social phobia. The proposed study is coordinated with the already-funded 12-week acute treatment study of PZ, CBGT, and PZ/CBGT. Responders and minimal responders from the acute study enter a 12- week intensive continuation phase to assess magnitude of further improvement. Responders at Week 24 enter a 28 week maintenance phase. Responders at Week 52 are then assessed for durability of response over one year of follow-up. Outcome is evaluated by blind assessor interviews, self- report questionnaires, and standardized behavioral tests. This study should substantially enhance our knowledge of effective long-term treatment for social phobia.